Validating Osteoporosis Detection in Rho AI Software

To assess the effectiveness of 16bit's Rho AI (artificial intelligence) software at identifying known cases of osteoporosis. 800 de-identified images from January 2007 to January 2024 will be accessed to test the software prospectively.

Osteoporosis is underrecognized and undertreated in the orthopedic surgery patient population. Osteoporosis is diagnosed by dual-energy x-ray absorptiometry which is costly and imparts radiation. Preoperative radiographs are universally obtained in orthopedic surgery patients, however up to this point radiographs have not been able to diagnose osteoporosis. This software program is a new technology that is able to leverage radiographs to opportunistically screen for osteoporosis. The study team will identify applicable patient records from medical record review, and Radius will de-identify the x-ray images. The de-identified images will be run through the Rho AI program to identify osteoporosis or osteopenia, and a report will be generated to summarize the program's findings.